68kMLA Classic Interface

This is a version of the 68kMLA forums for viewing on your favorite old mac. Visitors on modern platforms may prefer the main site.

Click here to select a new forum.
Screen stretch and neon lights….
Posted by: jimjimx on 2023-12-19 07:49:32
512ke has been perfect for 10 years, now this funkiness out of nowhere…. Is anybody able to give me a starting point? And what do the neon lights do? I’ve never seen them lit up before….
I assume it’s diverting current from somewhere?
Posted by: joshc on 2023-12-19 12:01:41
I would start with reflowing solder joints on the analog board - the yoke connector is often the culprit of weird behaviour like this,

The orange ‘light’ on the other hand is something I’ve never seen before. @cheesestraws is that something you’ve ever come across?
Posted by: Phipli on 2023-12-19 12:07:32
I would start with reflowing solder joints on the analog board - the yoke connector is often the culprit of weird behaviour like this,

The orange ‘light’ on the other hand is something I’ve never seen before. @cheesestraws is that something you’ve ever come across?
"The two neon bulbs act as voltage clamping devices to prevent serious overvoltages on the
filament and cathode terminals of the CRT. Similarly, the 1.5kV spark gaps integral with
C15-C17 act to protect the CRT under transient fault conditions."

https://vintageapple.org/gamba2/images/plus_analog.PDF

But I'm not sure, should it be continuously lit? Or is it indicative of a fault - defer to someone who actually has something this old or runs their early macs with the case off.
Posted by: cheesestraws on 2023-12-19 12:28:20
And what do the neon lights do? I’ve never seen them lit up before….
I assume it’s diverting current from somewhere?
The orange ‘light’ on the other hand is something I’ve never seen before. @cheesestraws is that something you’ve ever come across?

Slightly out of my comfort area here, treat this with that in mind please:

The neon bulbs - and they are neon bulbs - are overvoltage protection. The way that neon bulbs work is that they sit there totally inert until there's more than their strike voltage across them, then they light up, and it's significantly easier to keep them alight than it is to start them going. So you have a fairly robust, way to clamp the voltage across the neon to the voltage you need to maintain to keep it lit with a surge of its strike voltage.

Here's the circuit diagram from Thomas E. Lee's summary of how the AB works (the same document that @Phipli) linked above:

Screenshot 2023-12-19 at 20.12.21.png

And while I'm not going to pretend to be an expert in CRT circuitry, you can kind of see from this that the neons (NE1, NE2) are in a position where they can prevent dangerously high voltages reaching the CRT.

So: are both neons lit? I can't tell from the photos above. If only NE2 is lit, then that might point to something bad happening to the 30V feed, which is one of the taps off the flyback, and might actually point to a dodgy flyback.

(now go back and read my comment about being out of my comfort zone to understand the size of the error bars on this)

Or is it indicative of a fault - defer to someone who actually has something this old or runs their early macs with the case off.

It's indicative of a fairly big fault.
Posted by: jimjimx on 2023-12-19 13:09:15
I would start with reflowing solder joints on the analog board - the yoke connector is often the culprit of weird behaviour like this,

The orange ‘light’ on the other hand is something I’ve never seen before. @cheesestraws is that something you’ve ever come across?
Ive already done a reflow on the whole board years ago, and everything still looks good and shiny. Also replaced 3.9uf cap at C1.. Yoke connector is good too...
Posted by: jimjimx on 2023-12-19 13:13:06
So: are both neons lit? I can't tell from the photos above. If only NE2 is lit, then that might point to something bad happening to the 30V feed, which is one of the taps off the flyback, and might actually point to a dodgy flyback.
Only the bottom one is lit.. I can't tell which one that is, in the circuit...
Posted by: jimjimx on 2023-12-19 13:15:07
Also, the screen is dim, with the brightness turned all the way up.
Posted by: cheesestraws on 2023-12-19 13:42:08
The bottom one is NE2, which suggests an overvoltage on the 30V line to me: that comes straight out of the flyback.
Posted by: jimjimx on 2023-12-19 16:17:22
The bottom one is NE2, which suggests an overvoltage on the 30V line to me: that comes straight out of the flyback.
So, I guess I'm on the hunt for a flyback transformer?
Posted by: cheesestraws on 2023-12-20 02:03:06
So, I guess I'm on the hunt for a flyback transformer?

I suspect so - I can't see any other way that neon would light up - but as I said before please bear in mind I'm out of my comfort zone here, and I'm not going to state that with 100% certainty.
Posted by: Arbee on 2023-12-20 06:00:12
The flyback output going too high could be just too high of an input voltage due to a problem in the power supply section. Usually flyback failure results in no or low output voltage, although weirdness is always on the table there. Check the 12 volt rail to the horizontal circuit.
Posted by: cheesestraws on 2023-12-20 06:15:08
Check the 12 volt rail to the horizontal circuit.

Yes, this is a good plan - note, though, that the 12V rail on these is the one that's regulated, and all the other lines are derived from the 12V line. So if the 12V line was far enough out to cause this kind of overvoltage, I'd expect the 5V line to be so far out that the machine would be extremely unhappy - this wouldn't just show up as a video symptom. However, once again, only an opinion.
1